Join us at Microsoft Ignite 2026

Generalized model gains are flattening because the data that would improve a model on your domain sits inside your company. Discover how customers use Fireworks to customize open models and create differentiated AI experiences at lower cost. For architects and technical leaders weighing open models: Explore which workloads to consider, the training needed, and how shifting 2% of weights can propel an application past its category. Experience a case study and live demo of Fireworks on Foundry.

Enterprise token consumption has exploded, leaving engineering teams balancing AI innovation against rising costs. Learn how intelligent model routing for software development can help you do both. Discover how task- and cache-aware routing directs requests to the right open and closed models to maintain code quality while reducing token spend. See real coding workloads in action through a customer example and live demo of Fireworks Nexus.

Open models are only as good as the platform that runs them. In this session, see how Fireworks on Microsoft Foundry takes frontier open models from evaluation to production — better price-performance, workload-specific sizing, and enterprise-grade reliability on Azure, plus a first look at Fireworks training via the marketplace. Learn from Motif’s best practices for using open models on Fireworks on Foundry to build agentic workloads for their customers.

Hero Foundry Breakout Session featuring Fireworks